The purpose of this project is to carry out a systematic search of the genome for susceptibility genes for non-insulin dependent diabetes (NIDDM). The study will be carried out in Mexican American families currently enrolled in the San Antonio Family Diabetes Study (SAFADS), a project currently funded by NIDDKD (R01 DK42273, N. Stern, PI). To date, 432 individuals from 29 different families have been enrolled into the SAFADS; our recruitment target is 720 total individuals, and we anticipate achieving this target by June 1994. SAFADS families are ascertained on a low-income Mexican American NIDDM proband. Blood lymphocytes are currently being EBV transformed for SAFADS participants and cell lines have been/are being established to provide a renewable source of DNA. The current project will use SAFADS families as a resource for conducting a gene search. Our strategy will be first to seek preliminary evidence for linkage by typing markers on an initial set of 177 individuals from 23 families already enrolled and secondly, to verify suspected linkages by typing markers on additional family members from the initial pedigrees and by typing other SAFADS families. We will use approximately 250 highly polymorphic PCR-based microsatellite markers that span the entire genome at regular intervals and then use linkage analysis to search for linkage between these markers and NIDDM or its precursor traits (i.e., glucose, insulin, or C-peptide levels). Our simulations indicate that if NIDDM were inherited as a Mendelian dominant trait, we would have excellent power of detecting evidence for linkage on the initial set of pedigrees. As new markers are typed and added to the database, we will perform two- point linkage analyses to screen for linkage. Those markers with suggestive evidence for linkage will be used in combined segregation and linkage analyses and in multipoint analysis. If evidence for linkage persists, additional markers will be typed in that region to "zero in" on the NIDDM susceptibility gene(s). The phenotypes to be considered in these analyses are NIDDM, the quantitative traits, glucose, insulin, and C-peptide levels, and a risk score derived from a predictive model for NIDDM.